Skip to content
Agent Memory
Skill

/check-context

Accurately check current context window and token usage. Use when the user asks about context usage, token consumption, or when monitoring context levels.

BOOST
From plugin
zylos-core
1.2k14 skills
Install
$ npx -y skills add zylos-ai/zylos-core --skill check-context --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/check-context

Context preview

The summary Claude sees to decide when to auto-load this skill.

Accurately check current context window and token usage. Use when the user asks about context usage, token consumption, or when monitoring context levels.

SKILL.md

check-context.SKILL.md
name: check-context
description: Accurately check current context window and token usage. Use when the user asks about context usage, token consumption, or when monitoring context levels.
user-invocable: false

Check Context Skill

Check current context/token usage. The data source depends on the active runtime.

When to Use

  • When the user asks about context usage
  • When the user wants to know token consumption

How to Use

First, check the active runtime:

node -e "try { const c=JSON.parse(require('fs').readFileSync(require('path').join(process.env.HOME,'zylos/.zylos/config.json'),'utf8')); console.log(c.runtime||'claude'); } catch { console.log('claude'); }"

If Claude runtime

Read the statusLine data file (updated after every turn):

cat ~/zylos/activity-monitor/statusline.json

Report from the JSON:

  • **Context usage**: `context_window.used_percentage`% used, `context_window.remaining_percentage`% remaining
  • **Tokens**: `context_window.total_input_tokens` input, `context_window.total_output_tokens` output (window size: `context_window.context_window_size`)
  • **Session cost**: `cost.total_cost_usd`
  • **Model**: `model.display_name`

If Codex runtime

Read token usage from the most recently modified Codex JSONL session file:

node -e "
const fs=require('fs'),path=require('path');
const base=path.join(process.env.HOME,'.codex/sessions');
let best=null,bestMtime=0;
function walk(d,depth){if(depth>3)return;try{fs.readdirSync(d).forEach(f=>{const p=path.join(d,f);try{const s=fs.statSync(p);if(s.isDirectory())walk(p,depth+1);else if(f.startsWith('rollout-')&&f.endsWith('.jsonl')&&s.mtimeMs>bestMtime){bestMtime=s.mtimeMs;best=p;}}catch{}});}catch{}}
walk(base,0);
if(!best){console.log('No session found');process.exit(0);}
const lines=fs.readFileSync(best,'utf8').split('\n').filter(Boolean);
for(let i=lines.length-1;i>=0;i--){try{const j=JSON.parse(lines[i]);if(j.type==='event_msg'&&j.payload?.type==='token_count'&&j.payload.info.last_token_usage){const u=j.payload.info.last_token_usage.input_tokens;const c=j.payload.info.model_context_window||128000;console.log('used:'+u+' ceiling:'+c+' pct:'+Math.round(u/c*100)+'%');process.exit(0);}}catch{}}
console.log('No token_count event found');
"

Report:

  • **Context usage**: pct% used (used / ceiling tokens)
  • Report the pct value clearly so the user knows if rotation is needed (threshold: 75%)
Read more
Ships withzylos-core

Zylos (/ˈzaɪ.lɒs/ 赛洛丝) — Give your AI a life

Get the whole plugin

Other skills on zylos-core.